Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=mpin20

Public Integrity

ISSN: 1099-9922 (Print) 1558-0989 (Online) Journal homepage: www.tandfonline.com/journals/mpin20

Herbert Simon: "Is's" and "Oughts" After Sixty Years

Phillip O'Reilly

To cite this article: Phillip O'Reilly (2011) Herbert Simon: "Is's" and "Oughts" After Sixty Years, Public Integrity, 13:4, 371-384, DOI: 10.2753/PIN1099-9922130405

To link to this article: https://doi.org/10.2753/PIN1099-9922130405

Published online: 08 Dec 2014.

Submit your article to this journal

Article views: 1874

View related articles

Public Integrity, Fall 2011, vol. 13, no. 4, pp. 371–383. © 2011 ASPA. All rights reserved.

ISSN 1099-9922/2011 $9.50 + 0.00. DOI 10.2753/PIN1099-9922130405

Herbert Simon: “Is’s” and “Oughts” After Sixty Years PhilliP O’Reilly

Abstract

In his seminal book Administrative Behavior, Herbert Simon asserted that ad- ministrative decisions consist of facts and values, the former objective, the latter subjective. Simon argued that ethical propositions are, by their very nature, con- textual, relativistic, and based upon value judgments. The preeminence of Simon’s view of ethics has left scholars looking for a foundation upon which to build an ethical system for public administrators. This article evaluates the logic of Simon’s arguments and presents evidence that there are transcendent moral principles. The implications of transcendent morals and their application to public administrators and instructors are discussed.

Keywords: ethics, ethics in public administration

Few authors have had a greater impact on the theory of public administration than Herbert Simon. First published in 1945, his seminal book Administrative Behavior is still employed in organizational theory classes. Perhaps nowhere has Simon’s influence on the study of administrative behavior been more far-reaching than in the area of administrative ethics. Simon argues that ethical propositions are, by their very nature, contextual, that is, relativistic, and thus there are no foundations, akin to the laws of motion in the physical sciences, upon which to build an ethical framework.

Simon’s moral relativism has gained preeminence among public administra- tion scholars and is used by practitioners in formulating policy. However, if the advocates of Simon’s logical positivism hoped that its ascension would liberate administrators from the shackles of ethics, they have been sorely disappointed. Having rejected ethical standards as a myth of the unenlightened, administrative scholars still shy away from a headlong rush into moral relativism. Rather, they

372 • Public integRity fAll 2011

Phillip O’Reilly

seek principles that can guide public administrators through the ethical shoals of everyday decisions.

Academicians and, to a lesser extent, practitioners find themselves navigating be- tween the Scylla of moral relativism and the Charybdis of moral absolutes. They find comfort in the stability of ethical standards but reject the rigidities inherent in such a system. The liberation of moral relativism is enticing but its ambiguity disconcerting. Moral standards have been replaced with rule-based systems of ethics. Behavior that would have been molded by admonitions toward honesty and fiduciary obligation has been replaced by purchasing guidelines; self-control and modesty are replaced by sexual harassment policies. Although rule-based systems serve a purpose, they often venture to the extremes—they proscribe too much behavior or too little.

In some respects, the ascent of moral relativism is perplexing. This philosophy is not intellectually compelling and is often accepted at face value but ignored in practice. These issues will be addressed below by discussing (1) the basic premises of Simon’s argument, (2) the implications and shortcomings of his approach to ethics, (3) the strong evidentiary case for moral absolutes, and finally (4) the implications of transcendent moral standards for the discipline.

Simon’s View on ethics or Values

Simon (1997, 55) discusses ethics within the context of administrative decisions. Most policy decisions, he claims, embody objective and subjective components. The objective component includes such things as soil assessments and the engineering requirements imposed by the state of the soil when a dam is planned. The subjec- tive component, what Simon calls a “value element” or an “ought,” might include the deliberation over whether to dam a wild and scenic river. Objective or “factual” components are empirically testable and therefore may be correctly or incorrectly resolved. The same cannot be said for the oughts.

The value judgments or oughts embedded in decisions defy assessment. The policymaker confronted with conflicting oughts—one ought not to dam a wild and scenic river, versus one ought to promote hydroelectric power—will find no objec- tive criterion for judging between the two. In fact, the policymaker should expect no such criterion, because all assessments of this kind are relative. The veracity of an ought, according to this argument, always rests upon another ought. The chain of ethical arguments never leads to an objective standard. Rather, at its beginning the chain “must start with some ethical premise that is taken as ‘given’” (Simon 1997, 59). There is no right or correct decision in the sense that one makes the cor- rect decision about engineering techniques. Ethical values, in the final analysis, are simply a reflection of the preferences of an individual or a group.

Within this framework, administrators should make correct factual decisions but cannot expect to apply the same criterion to ethical ones. Simon states the difference between the factual and value components of decisions as follows:

the reason for making a division of this sort lies in the different criteria of “cor- rectness” that must be applied to the ethical and factual elements in a decision. “Cor- rectness” as applied to imperatives has meaning only in terms of subjective human values. “Correctness” as applied to factual propositions means objective, empirical truth. If two persons give different answers to a factual problem, both cannot be right. Not so with ethical problems. (1997, 62)

Herbert Simon: “Is’s” and “Oughts” After Sixty Years

Public integRity fAll 2011 • 373

Ideas or assertions that are empirically verifiable provide standards that no ethical principle affords its users. Ethical propositions, Simon claims, must be accepted on faith.

Assessing Simon’s View

The distinction between the empirical and the subjective finds many advocates in a society that elevates scientific knowledge and values open debate. However, if by empirical testing one means testing scientifically, that is, testing through the scientific process, Simon’s dichotomy between objective and subjective becomes strained. For in a very important sense all knowledge is based upon assumptions that cannot be verified empirically. These assumptions (the “oughts,” to use Simon’s vernacu- lar) must be taken as given. For example, all scientific tests assume continuity: The world ought to operate today as it operated yester- day and ought to operate tomorrow as it does today. This premise cannot be empirically tested but is accepted on the conviction that it is true. The assumption is accepted on faith; however, few would argue against its good sense. It provides a reasonable and essential foundation for all scientific tests, not because it can be empirically proven but because humans observe continuity in the world. Faith in the assumption is based upon evidence and reason.

One might argue that is’s and oughts remain distinct because scientific experiments can validate the assumption of continuity but cannot validate the basis of ethical imperatives. That is, the replication of experiments demonstrates continuity in nature and thus validates the assumption of continuity. No amount of experimentation can, however, validate ethical imperatives.

The flaw in this argument is that it confuses cause and effect. A scientist conducts experiments not to prove continuity, because replicating an experiment does not prove that continuity will exist in the future. Rather, when conducting an experi- ment a scientist relies upon the validity of the assumption of continuity. The very reason that experiments occur is that the scientist has a reason-based faith that the outcome will be replicable.1

Another, more esoteric example may prove helpful in assessing the validity of Simon’s distinction between the empirical and the subjective. A relatively small mi- nority of people claim that matter is illusory (England 1954, 450–451). One cannot empirically disprove this claim. It is possible that they are in fact correct. Why do more people not adopt this viewpoint? It is not because of any empirical test con- ducted to demonstrate that matter exists apart from the human imagination. Rather, individuals evaluate what they observe about the world and, more specifically, about how people behave. The most logical explanation of their observations is that humans exist in a material world. The proof is less akin to a scientific or mathematical proof than to the evidentiary proof that attorneys rely upon in a courtroom. The point is that the empirical tests provide tools for assessing the validity of facts. It does not follow from this statement that they are the only, or necessarily the preferred, tools for validating all facts.

If one is inclined to accept this argument, the door is opened to a much broader range of ideas which can then be characterized as objective, as facts: Abraham Lin-

In a very important sense all knowledge is based upon assumptions that cannot be verified empirically.

374 • Public integRity fAll 2011

Phillip O’Reilly

coln was the sixteenth president of the United States; the Nazi government murdered six million Jews; Mr. Smith had oatmeal for breakfast this morning. These ideas are verified as objective, as facts, or as is’s, through observation, testimony, and validating the reliability of the evidence.

The question now becomes whether this digression is of any assistance in as- sessing the reality or existence of moral absolutes. It is. The same techniques that verify the existence of matter prove valuable in assessing whether there are ethical standards that transcend human invention and preferences. To put it more loosely but in a manner that promotes brevity, these principles are valuable for assessing the existence of ethical imperatives.

the case for a transcendent Standard

At this point, it is appropriate to ask why those in public administration would be concerned with ethical codes in general and, in particular, with the existence of ethical

standards. This question is easily answered. Society needs a mechanism whereby it can assess whether public servants’ behavior is appropriate. Now if by “appropriate” one simply means preferable to some other be- havior, then the discussion can end at this point. Simon’s ethical structure is adequate for such assessments. However, most public administrators and most citizens have some- thing else in mind, for “appropriate” has a

moral or ethical connotation. Behavior conforming to certain moral principles is commended; behavior is condemned when it falls short. Appraising behavior within this ethical context requires some baseline or standard. From where does such a standard spring?

Some claim that moral codes spawn from the human survival instinct and a desire to promote society’s welfare. In this case, then, the role of the ethicist is to deter- mine what ethical structure best accomplishes these goals. However, theoreticians creating ethics ex nihilo inevitably encounter an insurmountable obstacle. They cannot satisfactorily demonstrate the superiority of their ethical code over alternative ethical codes. The ethicist stumbles precisely as Simon predicted. Moving along a chain of oughts until arrival at a premise that is taken as given, the ethicist can only assert that society should abide by this ethical code because it is preferable—but people’s preferences are based upon opinions and circumstances, and both change. Furthermore, assertions founded upon opinion carry very little weight; anyone is free to assert an alternative code based upon an alternative opinion. Under this construction, no metric for assessing alternative ethical systems exists. This is not a very satisfying solution from the theorist’s standpoint.

One might argue that although the theorist may be dissatisfied with this outcome, the lack of resolution leaves the practician unaffected. Cannot the practician focus solely on behavior and ignore more theoretical questions about ethical systems? The answer is no. The practician might assert that behavior that promotes society’s survival or welfare is commendable, and behavior that does not accomplish this criterion is unacceptable. This assertion simply pushes the question back another

Theoreticians creating ethics ex nihilo inevitably encounter an insurmountable obstacle. They cannot satisfactorily demonstrate the superiority of their ethical code over alternative ethical codes.

Herbert Simon: “Is’s” and “Oughts” After Sixty Years

Public integRity fAll 2011 • 375

level: Who determines whether behavior meets this criterion? The Nazis argued that their policy of discrimination and, later, genocide against the Jews promoted the survival of their society. Judging by the improvements in German productivity and their improved self-confidence after the rise of Hitler, one might contend that this policy satisfied the above criteria. Thus the practician is also left in a quandary. Since there is no satisfactory standard with which to accomplish the practical goal of assessing behavior, the practician is no better off than the theorist.

Neither the theorist nor the practician finds a passage out of this quagmire absent ethical standards that provide a foundation for moral assessments. If behavior is to be judged apart from human preferences, it requires moral standards that transcend human existence. Such standards would not be subject to the vagaries of human institutions and would be equally valid for assessing behavior within a culture and across cultures.

Regrettably, the fact that moral standards might come in handy does not guarantee their existence. Cold fusion would be beneficial but has so far proven elusive. The difference between cold fusion and a transcendent ethical standard, however, is that there is a strong case to be made that the latter does exist.

One category of evidence is the prevalence of consistent standards across many cultures and ages. Gilman and Lewis (1996) argue against the concept of cultur- ally specific morals. Using the term “fundamental values,” they assert that there is broad acceptance of honesty, fairness, and integrity across cultural and international boundaries (Gilman and Lewis 1996, 518). Their results provide mixed evidence (some of which is discussed below), but largely support their conclusion. Similarly, Yoder and Cooper (2005, 317–318) and Cooper (2004) argue that officials in vari- ous countries are committed to self-determination, honesty, and trust. Lewis (1955) offers a much longer perspective, documenting the consistent ethical teaching in ancient literature from both the Orient and the Occident.

Some contradictory evidence is presented. For example, Gilman and Lewis (1996) sampled residents of eleven international cities. They discovered that people believe in moral absolutes. In fact, majorities in seven cities believed that “clear guidelines about good and evil . . . always apply to everyone, whatever the circumstances” (520). However, a majority of respondents in four cities agreed with the statement that “what is good and evil depends entirely upon the circumstances at the time” (ibid.). Further, the authors mentioned above are quick to distinguish between preaching and practice. As Yoder and Cooper point out:

Whether these values are enacted or just espoused is not known and goes beyond our study. However, we argue that the fact they are espoused international values indicates they are at least viewed as aspirations. What we can say is that there may be an emerging consensus that these at least are the values which nations of the world believe they ought to say they support. The profession of them is significant. (2004, 399 [emphasis in original])

This distinction between principles and practice raises an important question. Do people simply claim they believe in “clear guidelines” because they believe it is expected of them, or do they behave as if moral standards impose constraints upon their behavior? This leads to a second category of evidence in support of transcendent moral standards, the evidence from human behavior.

Certain human behaviors are very difficult to explain in the absence of transcen-

376 • Public integRity fAll 2011

Phillip O’Reilly

dent moral standards. Observe someone who has been slighted in some way; for example, a person in a line when someone else cuts in front. In such instances, the slighted party commonly confronts the offending party with a statement that such behavior is unfair. The offending person rarely dismisses the charge on the basis of subscribing to a different moral code, but instead justifies the behavior by arguing that the code does not apply in this particular instance (Lewis 1952). The offender may appeal to a pressing engagement or to the fact that the person in line was also not abiding by the rules and moving forward in the line. Whatever the rationale, the point is that both people appeal to the same standard. Further, both agree that this standard sits in judgment on their actions, and both feel constrained by this standard.

One might assert that such behavior is unsophisticated and, thereby, inconclusive. Consider a subtler example and evidence from a hostile source. One of the arguments that is raised against the existence of transcendent moral standards relates to the rigidi- ties of such standards. Regardless of which ethical standards one codifies, there are situations in which one is forced to make exceptions to the standard (Harmon 2005). If one must make exceptions to the moral code, the argument goes, then the standard becomes a guideline, not an absolute, and one is back to moral relativism. This argument poses a problem for moral absolutism that will be addressed later; for the moment, consider the assumptions adopted by the moral relativist in making this argument.

There are two reasons why one might need to make exceptions to a moral code. The first is that abiding by the code is inconvenient and making an exception is ex- pedient. Moral absolutists and many relativists would agree that exceptions are not justified in such circumstances. There are more rare but more troubling situations when both the advocate and the critic of absolutes would claim that an exception to the rules is required. A real-world illustration may clarify the issues.

During the Nazi occupation of Holland, certain individuals, who subscribed resolutely to the moral imperative against lying, hid Jews from the Nazis. Assume you are such a person, you are hiding Jews in your basement, and Nazi soldiers enter your house and ask if there are any Jews present. Under these circumstances, the moral relativist would advocate lying. The reason lying is advocated over honesty is not primarily motivated by expedience.2 Most would advocate lying because they view it as the honorable thing to do. However, by justifying lying on the basis of honor, justice, or fairness, the relativist emulates the slighted person standing in line. The relativist in this case appeals to a standard but feels there is no need to defend because it is commonly known and accepted. The moral relativist believes it unfair to kill someone because of their ethnicity and further believes that lying is justified to avoid such atrocities.

The relativist argues, in essence, that people are sometimes morally constrained to violate moral absolutes; therefore, moral absolutes cannot exist. However, this argument is self-refuting. The relativist is judging between two moral principles, protecting the innocent and honesty. However, one cannot judge between alternative moral positions unless there is some standard by which both are measured (Lewis 1952, 13). By asserting that one course of action is morally superior to the other, the relativist is arguing that transcendent moral standards do not exist. Yet the relativist appeals to a transcendent standard to make this case. The moral relativist exhibits the same inconsistencies as a person who claims that matter is illusory—arguing from an intellectual standpoint that transcendent standards do not exist, but behaving in such a way as to provide evidence to the contrary.

Herbert Simon: “Is’s” and “Oughts” After Sixty Years

Public integRity fAll 2011 • 377

The position of the relativist as described above stands in stark opposition to the passage quoted from Administrative Behavior earlier. Simon, if one believes his statement, would argue that two persons, one choosing honesty and the other choosing to protect the innocent, would be on equal moral footing. That is, both would make “correct” decisions. If Simon’s assertion is true, his is a lone voice in the wilderness proclaiming that truth. Few people would agree that these are equally correct decisions. The relativist, in contrast, asserts that one decision is correct and one is incorrect. This is not what Simon argues.3

Returning to the issue at hand, the evidence indicates that humans behave as if transcendent moral standards exist. Furthermore, humans expect other humans to understand that such standards exist, and these expectations are reciprocal. This expectation was evident during the Nuremberg trials and is often reflected in the behavior of those who reject the existence of moral absolutes.

Even as rigorous a determinist as Karl Marx, who at times described the social be- havior of the bourgeoisie in terms which suggested a problem in social physics, could subject it at other times to a withering scorn which only the presupposition of moral responsibility could justify. (Niebuhr 1960, 79)

What conclusion is to be drawn from this? There appear to be only two explanations for this common human trait. Either human beings are uniformly deluded, or moral standards which transcend and, therefore, sit in judgment upon human behaviors do indeed exist. One is hard pressed to find evidence lending credence to the first explanation, which leaves the conclusion that moral absolutes are transcendent.4

Some, no doubt, find this conclusion disconcerting and will seek other explana- tions for this peculiarity of human beings. This is a necessary exercise, but should the other explanations prove impossible, then what remains? In the words of Sher- lock Holmes, “when you have excluded the impossible, whatever remains, however improbable, must be the truth” (Conan Doyle 1975, 179). University settings are not particularly amicable toward those who adopt the hypothesis that transcendent standards exist. Still the hypothesis should not be rejected simply because some are uncomfortable with it. If the hypothesis withstands attempts to refute it, honest people should rethink their positions.

Objections to a transcendent Standard

One argument against the existence of moral absolutes—the difficulty moral di- lemmas present to those advocating moral imperatives—has already been raised. However, before that objection is addressed, four others will be briefly discussed.

Some dismiss the existence of moral standards on the grounds that such standards cannot be ascertained. A transcendent standard requires revelation by a transcendent being. In the absence of revelation, the standards are of no use because they cannot be discerned. Obviously, this argument assumes that the moral standard has not been revealed. The observations already discussed cast significant doubt upon this assumption. People behave as if the standard or, at the very least, a good portion of it, has been revealed.

This objection is sometimes fashioned so as to cast doubt upon the existence of moral standards. However, such reasoning overlooks the distinction between

378 • Public integRity fAll 2011

Phillip O’Reilly

the existence of a phenomenon and our human ability to characterize it. Gravity existed long before Newton. He simply developed a systematic way of describing it. This is an important distinction, because if moral standards do exist, humans can employ tools of argumentation, logic, and observation to flesh out those standards and how they apply in public administration or, for that matter, any other area of life. This differs dramatically from the situation described by Simon in which no amount of debate, logic, or observation can identify the moral superiority of one action over another.

The second objection contends that moral absolutes find their genesis in other sources. Transcendent morals, the argument goes, are unnecessary to observe the lessons of history and discern which behaviors are advantageous to humans and which are destructive. While it is true that one can draw many lessons from history,

such observations do little to reconcile the debate over which historical figures should be emulated. Typically, discussions of histori- cal figures involve some moral connotations. However, history itself provides no tools for ranking behavior on an ethical scale. To ac- complish that task, one is again confronted with the need for a moral code. Attempts to attribute morals to instinct, social norms, or other human sources run into the same obstacles. When circumstances require that one identify which instinct or norm is mor-

ally preferable, instincts and norms provide no tools for accomplishing this task (Lewis 1952, 10).

A third objection comes from the fields of neurology and psychology, which Macaulay (2009) and Harmon (2005) summarize in nonprofessional terms. Objec- tions in this category take several forms but contain a common thread. Each claims that physiological traits, environmental stimuli, or some combination thereof trump ethical standards as an influence on behavior. Evidence from these fields could be used as proof for two conclusions: first, that transcendent moral principles do not exist, and second, that the influence of moral principles is limited. For the first conclusion to be true, all cognitive processes and behavior must be a response to environmental stimuli by a person with certain physiological predispositions. In such a world, there would be neither free will nor decisions with moral implica- tions; there would be only stimulus-response. This proposition, however, is logically inconsistent. Assume that all behavior is strictly a response to natural stimuli. How would one ascertain this fact? One could not discover it from within the stimulus- response cycle, for any stimulus determines a response, not some observation of a response. Analyzing the behavior would require transcending the stimulus-response cycle. The very act of advocating s